Year: 2024
Type: article
Abstract: This methods paper describes the methodological and statistical underpinnings of the highly adaptive testing design (HAT), which was developed for the Programme for International Student Assessment (PISA). The aim of HAT is to allow for a maximum of adaptivity in selecting items while taking the constraints of PISA into account with appropriate computer algorithms. HAT combines established methods from the area of computerized adaptive testing (a) to improve item selection when items are nested in units, (b) to make use of the correlation between the dimensions measured, (c) to efficiently accomplish constraint management, (d) to control for item position effects, and (e) to foster students' test-taking experience. The algorithm is implemented using the programming language R and readers are provided with the necessary code. This should facilitate future implementations of the HAT design and inspire other adaptive testing designs that aim to maximize adaptivity while meeting constraints.
Source: Frontiers in Psychology
Authors: Aron Fink, Christoph König, Andreas Frey
Institution: Goethe University Frankfurt
Language: English
Field: Decision Sciences
Domain: Social Sciences
SDG: Quality education
Open Access status: gold
Estimated APC paid: $3,813
OpenAlex ID: W4402604674